constructing equality? women’s wages for physical labor, 1550...
TRANSCRIPT
LUND UNIVERSITY
PO Box 117221 00 Lund+46 46-222 00 00
Constructing equality?
Women’s wages for physical labor, 1550-1759Gary, Kathryn E.
2017
Document Version:Publisher's PDF, also known as Version of record
Link to publication
Citation for published version (APA):Gary, K. E. (2017). Constructing equality? Women’s wages for physical labor, 1550-1759. (Lund Papers inEconomic History; No. 158). Lund: Department of Economic History, Lund University.
General rightsUnless other specific re-use rights are stated the following general rights apply:Copyright and moral rights for the publications made accessible in the public portal are retained by the authorsand/or other copyright owners and it is a condition of accessing publications that users recognise and abide by thelegal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private studyor research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal
Read more about Creative commons licenses: https://creativecommons.org/licenses/Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will removeaccess to the work immediately and investigate your claim.
Lund Papers in
Economic History
No. 158, 2017 Education and the Labour Market
Constructing equality?
Women’s wages for physical labor,
1550-1759
Kathryn E. Gary
DEPARTMENT OF ECONOMIC HISTORY, LUND UNIVERSITY
Lund Papers in Economic History
ISRN LUSADG-SAEH-P--17/158--SE+22
© The author(s), 2017
Orders of printed single back issues (no. 1-65) Department of Economic History, Lund University
Postal address: P.O. Box 7083, S-220 07 Lund, Sweden
Telephone: +46 46 2227475
Telefax: +46 46 131585
Full-text electronic issues (no. 58, 60, 61, 66--) www.ekh.lu.se
Lund Papers in Economic History are published by the Department of Economic History,
Lund University, Sweden. This series replaces the former series under the title Meddelande
från ekonomisk-historiska institutionen, Lunds universitet. The change of name reflects the
orientation of the series towards an international readership. The series is multilingual, but the
majority of the working papers appear in English.
Lund Papers in Economic History include papers in the following topic areas:
General Issues
Development Economics
Education and the Labour Market
Population Economics
Lund Papers in Economic History are published as occasion arises, not at fixed intervals.
Printed issues are distributed to libraries. From 1999 and onwards, full-text electronic issues
are also available on www.ekh.lu.se. Those who would be interested in receiving information
by email on new issues of Lund Papers in Economic History are requested to send an email
message to [email protected].
Constructing equality? Women’s wages for
physical labor, 1550-1759
Kathryn E. Gary
Lund University
Department of Economic History
Abstract
This paper combines new archival data on women’s wages from southern Sweden with
published series from Stockholm in order to create a series of early modern female
construction workers’ wages between the middle of the sixteenth and middle of the eighteenth
centuries. This paper finds that women had relatively high relative wages in the later part of
the sixteenth century, with an increasing wage gap into the eighteenth century, and that the
changes in women’s relative remuneration are connected to changes in demand factors.
This paper challenges assumptions about women’s participation in manual labor, in many
cases finding a lack of differentiation between female and male unskilled workers as well as
and unskilled labor force comprised of from forty to sixty percent women and high work
intensity for female construction workers.
Keywords: Gender wage gap; wages; women; Scandinavia; Sweden; early modern period.
2
K. E. Gary
Introduction
Investigations of historical wages and the development of living standards in the early modern
period have been increasingly popular, especially since Robert Allen’s 2001 paper on the
development of real wages throughout Europe. However, these wage studies have almost
universally referred to men’s wages as the drivers of economic development and determinants
of living standards; comparatively there has been less investigation into women’s earnings or
into the ways in which they participated in paid work. But with strong evidence that the
household and family-level division of paid market labor changed significantly over the early
modern period (de Vries 2008), it is essential to understand women’s contributions more
clearly if we want to understand living standards more completely.
It has long been taken for granted that early modern women did not participate in paid work in
the same ways as men; they have been assumed to have been constrained to unpaid and
domestic labor or to cottage industry, with their participation in casual work, especially
physical labor, largely overlooked. Recent research has increasingly indicated that unskilled
women often participated in paid market work in similar ways to unskilled men, patching
together a full-year’s work schedule from short-term jobs as they became available, working
for different employers and across different sectors (Humphries and Sarasua 2012), but there
is still little known about women’s earning and labor patterns in the early modern period.
This study fills some of these gaps by developing a series of women’s wages from the
building sector in Sweden, combining this new data from the south with an already-published
series of female builders’ wages from Stockholm, to create a series of Swedish women’s real
and relative wages from the middle of the sixteenth century through the middle of the
eighteenth. Together with men’s wages from the same sources, this provides new insight into
the relationship of men’s and women’s wage trends. The scope and detail of the data combine
to provide both a macro view of wage development in for women and men in unskilled
construction in Sweden, as well as a micro level perspective on individual and gender-based
work patterns.
This study addresses the following research questions:
1. How did the wages of unskilled female and male builders develop in early modern
Sweden?
2. Did women’s wages in the early modern Swedish construction industry gain relative
to men’s, decrease relative to men’s, or were they at a consistent ratio?
3. What are the mechanisms which drive relative wage developments? Is there evidence
for productivity-biased payments, or for women receiving a customary percentage of
men’s wage rates?
4. How did women’s participation in paid manual labor compare to men’s? To what
extent did their employment differ?
Preliminary findings challenge long-held beliefs about women in physical labor; in the end of
the sixteenth century female construction workers in Sweden were able to earn wages close
to, or on par with, their unskilled male coworkers, though women’s relative wages declined in
3
K. E. Gary
the seventeenth century, and women were of course typically relegated to the least skilled
construction jobs. Women’s relative remuneration appears to be largely demand-driven, with
higher relative wages in periods with more construction work; there is less evidence for
systematically discriminatory or custom-based wages; during peak building periods the labor
force was especially feminized, with women constituting between 30 to 60 percent of the
unskilled labor force during these periods.
Previous research
Women’s work in the past has been systematically less studied than men’s. Some of this is
because of the relative ease of finding data – women are often not as clear in the historical
record, and their work has often been less formal, making them harder to collect or identify
(Humphries and Sarasua 2012). However difficulty does not mean that there is no data, and
the focus on women has been increasing.
Humphries and Weisdorf (2015) present a comprehensive, long term series of women’s wages
for both day and annual labor from the middle of the thirteenth century through the
nineteenth. Built on an impressive array of primary and secondary sources and covering a
wide range of types of work, this is the first long-term investigation of women’s wages in
early modern and late medieval Britain. They conclude that the ‘Golden Age of the Peasantry’
following the Black Death – typically associated with increased bargaining power and wages
for unskilled laborers – likely had little benefit for women, and that women who were
engaged in casual work were likely increasingly dependent on their husbands for financial
support as the gender pay gap widened through the period of industrialization; however, those
women who could commit to full time labor with an annual contract, typically young single
women without family commitments, had more ability to support themselves through the
early modern period.
The majority of gender wage gap studies in a pre-industrial context examine the gender wage
gap in the long nineteenth century; this literature tends to have a British focus, and to rely on
agricultural data collected from farm accounts. These studies haven’t typically produced long-
term national series of data, but instead look closely at a particular farm or set of farms in
order to gain a fuller picture of the working environment and division of labor between the
sexes. Burnette (2004) expands on this genre to examine a larger set of farms throughout
England from 1740 to 1850, also finding a decreasing female relative wage alongside a
decreasing demand for female farm labor over this period; she attributes the decrease in
demand for female farm labor to women’s increasing ability to find alternative employment in
cottage industry, driving up the competitive wage to the point where the wage farmers would
have needed to pay to hire women would have been higher than the value of women’s
productivity.
Several other studies have found a widening gender wage gap through the late medieval and
early modern periods. Van Zanden (2011) investigates the development of women’s unskilled
wages relative to men’s at Winchester College in England, finding that initial wage parity in
4
K. E. Gary
the early sixteenth century rapidly declined, with women earning less than half of men’s
wages by the middle of the seventeenth century. The decline in women’s relative wages is
also found in wage assessments – the maximum wage rates allowed for certain types of
workers – for reapers and haymakers across England in the early modern era; here the decline
in parity occurs later, and is not as drastic, as the actual wages from Winchester College. The
wage gap in the Netherlands follows similar patterns, with a very small wage differences
before the Black Death and no wage gap in observations in the first part of the sixteenth
century followed by a loss of wage parity going into the eighteenth century, though this
increase in inequality was more modest and occurred later than in England (van Zanden
2011). Though I have been unable to locate John Langdon’s 2010 conference paper
discussing the wages of female builders in the medieval period, van Zanden (2011) reports
results from this paper which also follow the trend of a widening gender wage gap into the
early modern period; Langdon finds a much higher wage ratio, nearing two thirds, in the
1290s than in the early 14th century, when wage ratios have decreased to only about three
quarters. At times, the wage ratio could approach levels as high as 80 to 100%.
Though there is increasing empirical evidence about women’s wages and the development of
gendered wage gaps, there are still difficulties in explaining what factors are driving
differences in payment rates. The primary debate centers on whether wage differences were
due to discrimination or whether they have been based on fair market forces, driven by the
differences in women’s and men’s productivity. The historiography surrounding the Black
Death also brings women’s response to supply and demand shocks into focus; theoretically,
women would be more incentivized to join the labor force when labor was in low supply, and
subsequently would be able to demand higher wages.
Much of the long-term decline in women’s relative wages has been connected to the labor
shortage following the Black Death, a ‘golden age’ for the peasantry and unskilled in general;
the low supply of unskilled labor in combination with continued and inflexible demand for
agricultural workers allowed unskilled laborers, and theoretically women in particular, to
demand higher wages; the subsequent decline in relative wages is connected to population
recovery in the following centuries (van Zanden 2011). However, Humphries and Weisdorf
(2015) find no such effect of higher women’s relative wages following the Black Death.
Bardsley (1999) criticizes the historiography that claims that women were paid on par with
men following the Black Death; she examines women’s relative harvest wages before and
after the Black Death in England, and finds a highly segregated wage structure, wherein the
best paid women could earn only as much as the lowest paid men, but average wages were far
from equal even when women and men were employed on the same tasks. She also presents
evidence that women’s wages may have been overestimated in some previous literature,
especially when women were performing as part of a work group, in which a lump payment
was made to a group leader and distributed internally. Further, many of the lowest male wages
were possibly paid to boys, older men, or less physically fit men, meaning that the pay gap
between fit adult women and fit adult men would have been even larger than it appears. Her
overall conclusion is that gender, along with age and perceived physical ability, was an
intrinsic component of how wages were determined throughout England’s late medieval and
5
K. E. Gary
early modern period; she stresses the likelihood that selective comparisons and the lack of
strong or systematic data has led to an overestimation of women’s early modern wage parity.
Burnette (2008) counters that women were not paid lower wages than men simply on the basis
of sex, and argues that when differences in productivity are taken into account women were
paid on par with men for manual work – wage differences are because women are not as
strong as men, and so are not as productive in strength-based tasks. While men and women
often completed different work, this sorting was market-motivated and led to greater
economic efficiency, rather than being based on discrimination or gender roles; while there
was some assumption of typical ‘male’ or ‘female’ tasks, workers and employers were almost
always willing to override gendered work categories in the interest of economic efficiency.
Burnette (2004) connects the decline of women’s relative wages in the eighteenth to
nineteenth century Britain to changes in farming practices and technology: as the labor-grain
price ratio shifted from low grain and high labor prices to high grain and low labor costs,
farmers preferred to pay the higher male wages, which allowed a higher reliance on the
heavier scythe and led to more efficient grain collection than with the sickle, which was more
easily used by women. Similarly, van Zanden (2011) connects the decline of wage parity to
the development of large, capital intensive farms and enclosure systems; as farms grew in size
and family farms disappeared, the demand for farm labor fell across the board, but fell
especially for women and children.
Apart from wages, women’s participation in paid work is also not well understood.
Humphries and Sarasua (2012) propose that women participated in the paid labor market far
more than what has historically been assumed, and that women were often occupied year
round with paid work though it might be comprised of several jobs across different sectors;
men’s work was much the same. The absence of women from the official record of work is
largely born of our own and our predecessors’ ideologies. They also argue that demand-side
factors were typically a stronger influence on women’s decision to work in the early modern
period than supply-side factors such as marriage or fertility, as de Moor and van Zanden
(2010) and van Zanden (2011) would suggest through a mechanism such as the European
Marriage Pattern; van Zanden connects this decline in women’s relative earning power to the
decline in women’s marriage ages that was occurring in the eighteenth century. This is in turn
theoretically connected to Becker’s theory of lower marriage ages being associated with
greater specialization within marriage, and falls in line with de Vries’ (1994) theories of the
Industrious Revolution and changing patterns of within-family specialization and the rise of
the bread-winner model.
Data
The primary data for this study are collected from payment records from institutional and
manorial archives in the south of modern day Sweden, primarily from the cities of Malmo and
Kalmar, though also from some manorial sources and smaller towns around the southernmost
province, Scania, where Malmo is located. All of the data in this current work represent day
6
K. E. Gary
laborers working in the construction industry. Johan Söderberg has generously granted access
to some of his archival notes used for the construction of the 1991 series; some of this
unpublished material is used to extend the series of women’s wages in Stockholm earlier in
time.
In most instances the data is coded using the HISCO and HISCLASS systems (Leeuwen et al
2000; Leeuwen and Maas, 2011). The exceptions are Kalmar, where the data is presented in
the original source already divided into skilled and unskilled labor, and the already-published
data from Stockholm. These systems allow for systematic categorizations of work tasks and
the differentiation of skilled (in the case of these data, medium skilled) and unskilled workers.
In primary data women are less frequently represented than men. Unskilled men are paid for
over 63,000 days of work 1550-1759; in the same period women are paid for 7324.
A female worker can be identified either through a gendered occupational title, a female
relationship indicator, or by her name. In some sources, such as Kalmar, virtually every
individual is named, which not only makes identifying the gender of the worker
straightforward but means that we can be confident that we are not misidentifying female
workers as male.
Women are not systematically separated from men through their titles or recorded
occupations. In Malmö, women are sometimes identified as a digger’s wife (grävarens
hustru), but other times they are called simply a digger (grävare), the same title given to men.
In Kalmar all unskilled workers are subsumed under the title ‘hantlangare’, or unskilled
worker.
Wages for unskilled men and unskilled women from Stockholm are taken from Jansson,
Andersson Palm, and Söderberg 1991.
Methodology
Annual daily wages are calculated as the unweighted mean of daily payments. Each
individual payment is given equal weight; a person who is recorded as receiving a payment
five times in one year is counted toward the average five times. This is to give the strongest
weight to the most typical wage, and because there are many times when it is impossible to
determine the identity of the individual. Heavy outliers are removed.
Real wages are computed by deflating the nominal annual series by a cost of living series. In
this case, the price basket is modeled after Robert Allen’s (2001) consumption baskets,
modified for a Swedish consumer. Day wages are multiplied by 250 days to represent an
annual income. Cost-of-living baskets represent the total costs of supporting a family of 3.25
adult male equivalents, to proxy a family of one man, one woman, and two children for an
entire year following Allen and Weisdorf (2011). The y-axis indicates how many times over
an adult’s wage can meet the financial needs of the family unit. A value of 1 indicates that an
adult’s wage could just support the family unit – a value under 1 indicates that the single wage
7
K. E. Gary
would not have been enough to support the family, while a value of 2 indicates that there is
enough income to support the family unit two times over. This analysis uses so-called
subsistence baskets – that is, a basket of goods that comprises cheaper and less luxurious
alternatives – in order to give a lower-bound look at the possible range of living standards. 1
Women’s relative wages are computed by dividing women’s daily unskilled wages by men’s
daily unskilled wages. A value of 1 indicates that men and women had the same daily wage
and a level of 0.8 indicates that women earned eighty percent what men did. To avoid
regional price effects relative wages are computed locally and then an unweighted average is
taken across all regions.
Women’s and men’s nominal and real wages in Scandinavia
Figure 1 shows a trend of fairly stagnant real wages for men; a mild increase in living
standards during the sixteenth and into the seventeenth century is followed predominantly by
stagnation for men through the rest of the period. After a slight increase in wages for women
going into the seventeenth century, women suffer a wage decline through the rest of the
1Consumption baskets are developed separately for Sweden in Gary (2017)
Figure 1: Unskilled men’s and women’s real wages in Sweden 1550-1759– in subsistence
baskets
02
46
8
1550 1600 1650 1700 1750year
Men's real wages Women's real wages
8
K. E. Gary
period. Income levels are typically between 1 and 2 subsistence baskets for both women and
men, though sometimes dip below; according to Allen’s methodology this indicates that while
unskilled Swedish workers could make ends meet when work was available, they were not
living luxurious lives. This is especially true as the subsistence basket measurements given
here are for the most basic level of survival, without allowance for extras.
It is important to notice the parallel movement of women’s and men’s real wages through the
first part of the seventeenth century, followed by decided wage divergence. There is some
catchup of women’s wages in the eighteenth century, but the data is scarcer here, so the trend
is not as conclusive.
Nominal wages (figure 2) show that women’s real wage decline is due largely to the halting
of women’s nominal wage growth during the majority of the seventeenth century and into the
eighteenth, while men’s nominal wages continued to keep pace with prices.
The development of men’s wages is largely in line with the trends outlined in Allen (2001),
which show a divide in growth between the developing Northwest and the poor and stagnant
‘rest’ of Europe and outlines the Little Divergence; Sweden compares with growth trends in
the stagnant periphery, which were left behind by wealthier economies in England and the
Netherlands.
The stagnation and decline of Swedish real wages can be partially attributed to the general
instability and turbulence of the seventeenth century, during which time Sweden’s position as
one of the great European powers led to almost constant warfare, most of it overseas, taking
substantial tolls on Sweden’s economic and demographic development.
Figure 2: Men’s and women’s nominal wages in SEK
Source: see text
0
0,05
0,1
0,15
0,2
0,25
0,3
0,35
0,4
0,45
15
50
15
57
15
64
15
71
15
78
15
85
15
92
15
99
16
06
16
13
16
20
16
27
16
34
16
41
16
48
16
55
16
62
16
69
16
76
16
83
16
90
16
97
17
04
17
11
17
18
17
25
17
32
17
39
17
46
17
53
Men's nominal day wages (SEK) Women's nominal day wages (SEK)
9
K. E. Gary
Women’s relative wages in Sweden
Figure 3 shows the women’s relative wages as a percentage of men’s in Sweden,
disaggregated by region. The dominant trend is an increase in women’s relative earnings into
the beginning of the seventeenth century, followed by a steady decline through the
seventeenth century; this is followed by a slight indication of a recovery at the beginning of
the eighteenth century, but this is tentative at best.
It is especially interesting that several data points from Kalmar and Stockholm indicate that
unskilled were at times able to out-earn unskilled men. Some of the high levels of women’s
relative wages can be directly connected to events that would both increase the demand for
building labor and likely decrease the supply of able men. Kalmar was previously a border
town between Sweden and Denmark; the Kalmar War was fought between these two
countries from 1611-1613, and the high wage ratio of 1.2 in 1614 is likely in response to the
need to rebuild coupled with a fall in the supply of men.
Peaks in women’s relative earnings in the late 1640s coincide with the years directly
following a fire that devastated Kalmar in 1647; the town was rebuild at a slightly removed
location, and building labor would certainly have been in high demand, helping to explain the
wage ratio of 1.33 in that year
Figure 3: Women’s relative wages as a percent of men’s wages, disaggregated by source
Source: see text
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
15
50
15
56
15
62
15
68
15
74
15
80
15
86
15
92
15
98
16
04
16
10
16
16
16
22
16
28
16
34
16
40
16
46
16
52
16
58
16
64
16
70
16
76
16
82
16
88
16
94
17
00
17
06
17
12
17
18
17
24
17
30
17
36
17
42
17
48
Malmö Kalmar Stockholm (archival)
Stockholm (published) Stockholm (Söderberg)
10
K. E. Gary
The high levels of women’s wages could also be connected to gender-specialization in the
unskilled (or low-skilled) construction workforce. When Kalmar was rebuilt following the
fire, burgers were given tax credits for building in stone, instead of wood. Throughout the
archival records carpenters typically work alone, without assistants. Masons on the other hand
often had many people working under them. In particular, they have more specialized
unskilled workers, called kalkslarare, or the feminine kalkslagerskor, who frequently
appeared with the same mason regularly. This position was often held by women, especially
in Kalmar. Given this pattern of professional relationships, and increase in the demand for
skilled male mason labor would also drive up the demand for semi-specialized unskilled
female labor.
Even given the unusual circumstances that help explain some instances of very high relative
wages, the decrease in women’s relative payment through the seventeenth century is dramatic,
especially as it occurs over the course of only a generation or so. The following sections
concerning women’s participation in the labor force aims to reveal some of the mechanisms
that led to this dramatic swing in women’s relative compensation.
Women as a part of the labor force
Demand-pull on women’s work
Women played an important role as paid manual laborers; this was especially the case in
periods when construction labor was in high demand. The importance of women in the
building industry is particularly evident in Kalmar: one of the unique features of this data is
that virtually every individual is named, which means that identification of each worker is
essentially universal. This allows for a examination of women’s presence in construction
field on an individual level. Kalmar was not a large town; it had only about 2000 inhabitants
around 1700 and the surrounding countryside was sparsely populated (Andersson Palm 2000),
but this was fairly normal for a Swedish town in the period. Unfortunately the population
statistics are not detailed enough for a clear examination of the population development of the
town; because of this, little is known about the supply side of labor in Kalmar or how it might
have changed in different periods.
11
K. E. Gary
Table 1 shows the number of payments made to women as a percentage of payments to all
unskilled workers in Kalmar and as a percentage of the total sample, presented with the
number of all unskilled workers and number of workers in total. Clearly the 1620s and 1630s
were the biggest years for construction in Kalmar, while construction work recorded in the
available sources generally declined in the 1640s; this is likely due to the extensive repair
work necessary after the church and other parts of the city were destroyed during the Kalmar
war 1611–1613. Women are a substantial part of the labor force throughout the period; they
are regularly about half of the unskilled labor pool, and even when skilled (male) labor is
included they are still a strong proportion of all builders. There is no doubt that female
building labor was an integral feature of construction work in Kalmar.
Table 1: Women’s work days as a percentage of
all workdays in Kalmar, 1614-1706. Workdays
rounded to nearest full day.
As % of
unskilled
workdays
unskilled
workdays
(n)
As % of
total
workdays
Total
workdays
(n)
1614- 1620 68.6 2502 39.0 4406
1621- 1625 31.8 4039 18.6 6903
1626- 1630 60.8 2907 52.7 3354
1631- 1635 69.4 762 54.1 977
1636- 1640 54.5 130 11.5 607
1641- 1645 40.3 144 38.4 151
1646- 1650 18.9 354 4.2 1601
1651- 1655 13.2 46 6.2 97
1656- 1660 0 69 0 73
1661- 1665 0 13 0 20
1666- 1670 0 27 0 130
1671- 1675 17.4 383 15.2 438
1676- 1680 100 13 52.0 25
1681- 1685 27.8 18 15.4 33
1686- 1690 20.0 175 16.3 215
1691- 1695 5.7 35 4.1 49
1696- 1700 4.1 122 4.0 126
1701- 1705 0 135 0 135
1706- 1710 0 13 0 51
>60% women
30-59 % women
Top quartile (n) of
workdays
12
K. E. Gary
The data here represents only one source from Kalmar, as others do not survive, but there are
indications that this source is a strong representation of the building sector in Kalmar. After
the fire in 1648, Kalmar was rebuilt at a new location with tax relief offered to burgers who
rebuilt their homes in stone. As mentioned above, female construction workers frequently
assisted masons, and so we can assume that female unskilled labor for private projects was in
high demand in these years along with building labor in general. Because this employment
would have been in the private sector, paid by individual townspeople, it does not appear in
the records of the public sources used in this study. This demand from the private sector was
enough to drive reallocation of the labor supply from the public market – at least, there are
significantly fewer payments to construction workers in the public record during the period
when private construction would have been at its highest. The few wages that do remain
continue at a high level, especially for women. This indicates that the Kalmar market is small
enough that the patterns in the records analyzed here are a reasonable proxy for the
construction market, and reflect workers’ reactions to supply and demand forces. This high
labor demand, combined with a limited labor supply, may also help explain women’s
relatively high wages in this decade, despite this decade having fewer recorded wages overall.
The high proportion of women working during years in which the labor demand was greatest
further indicates that women responded to demand factors when entering the labor market.
We can investigate women’s incentives further by examining the relationship between
women’s wages and labor demand.
Figure 4 shows the development of total labor demand, measured as the number of individual
wage payments made in each decade, along with women’s relative wages. It is important to
note that this comparison is somewhat flawed; the data of women’s relative wages contain
information from Stockholm, while the labor demand data do not. However, from figure 2
above we can see that the overall shape of women’s wage development is fairly consistent
even without the inclusion of Stockholm; the series would have more gaps and a slightly
higher level, but the shape would be stable.
As figure 4 shows, the association between building demand and women’s relative wages is
high; the correlations confirm this association; total individual days worked and women’s
relative wages have a correlation coefficient of 0.31, which indicates that either employers
raise women’s wages, though not men’s, in times of high labor needs, or that women are able
to enact some degree of collective bargaining when employed in larger groups. Together this
is strong evidence that demand factors have a strong influence on Swedish women’s
connection to the building industry, in accordance with Humphries and Sarasua’s (2012)
suggestion about the motivations of preindustrial working women in general.
13
K. E. Gary
A closer look at the difference in individual working patterns between men and women in
Kalmar, with its more-complete records, gives a better idea of how women’s work compared
to men’s. Notably we find one individual, a woman called Anna Dikerska, is an abnormally
tenacious worker. Weekly payment records record unskilled men and women together; this
list is organized in order of days worked in the week, with those who worked the most days
topping the list. This position is almost unwaveringly held by Anna Dikerska. Because she is
clearly unusual as an unskilled laborer, summary statistics are presented with and without her
better understand what is typical of women performing this labor.
Table 2 below shows the average number of days an individual worked in a year, the average
number of times an individual appears recorded as receiving payment (on a weekly basis), and
the average number of years over which an individual worked, although these years need not
be consecutive. Skilled men worked the most, which is to be expected, as they invested
considerable amounts of resources to learn their trades and pursue a livelihood. What is less
expected is that women, while fewer in number than unskilled men, appear to work more than
their male counterparts in an industry that relies on physical strength and necessitates absence
from the home. Median values are higher for unskilled women in all categories than for
unskilled men; women are on average working more days in a year, and are returning to the
workplace more often and over a longer period of time than unskilled men. While the raw
mean might not be higher in all categories – unskilled men work during an average of 1.39
years, while women work during an average of 1.36 – in combination with median and max
Figure 4: Total building demand in nuber of days worked (right axis)and women’s relative
wages in Sweden (left axis) by decade
Source: see text
0
2000
4000
6000
8000
10000
12000
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
Total paid work days (right axis) Women's relative wage (left axis)
14
K. E. Gary
values the data shows a clear tendency toward a higher work intensity from women than from
unskilled men. This possibly indicates that unskilled men had more options than women for
earning an income on a casual basis, while women were more restricted in their work
opportunities, but it is clear that women were a regular presence in the construction industry.
Table 2: Summary statistics of men’s and women’s work in Kalmar construction
Average annual workdays
mean s.d. min median max n
unskilled men 8.87 14.05 0.5 5 143 377
women 15.7 15.7 1 7 141 243
women w/o A.D. 11.35 15.73 1 6 114 242
skilled men 31.82 36.92 0.5 14 144 99
appearances (individual recorded payments, on the weekly level)
mean s.d. min median max n
unskilled men 3.5 5.86 1 1 74 377
women 5.28 15.98 1 3 226 243
women w/o A.D. 4.36 7.31 1 3 69 242
skilled men 13.79 27.40 1 4 186 99
Number of years over which individuals worked (need not be consecutive)
mean s.d. min median max n
unskilled men 1.39 0.95 1 1 9 377
women 1.36 1.29 1 1 14 243
women w/o A.D. 1.31 1.0 1 1 10 242
skilled men 1.65 2.0 1 1 18 99
Qualitative evidence from Southern Sweden
Evidence from Malmo also supports the idea that women tended to play a fairly regular role
in construction work; while probably less common than in Kalmar, women’s paid work seems
to have been fairly normal and to have gone without comment. One piece of evidence for this
is the frequency with which women are not given a gendered occupational title, but are only
identifiable by their name; this at least implies that women’s work was not seen as different
enough to record them as doing separate work from men. Additionally, whether or not women
were given a gendered title seems to have been somewhat arbitrary; Karine, wife of Niels the
digger (possibly Niels Graffuer), appears in the data at least four times, and possibly a fifth,
between 1589 and 1593. However, she is never named in the same way twice; she is named as
Nils’ Karine (Karine Nilssis), and as Niels’ woman Karine (Niels kvinna Karine), and her
occupation also changes from entry to entry. Twice she is listed as just a digger (grävare),
15
K. E. Gary
while twice she appears as a digger’s wife (grävares kvinna). In 1592, another Karine, Karine
Wogns, is paid 12 skillings for 1.5 days work on what was most likely the foundation work
for Malmo castle’s east tower. She appears directly after Wogn Jensön, her husband, who
receives 8 skillings for one day’s work, the same per-day rate as Karine. For both Karines,
neither their occupational titles nor their wage necessarily identify them as female. It seems
very likely that there are more women who similarly disappear into the data because it is
impossible from the data to see that they are women. This is consistent with Humphries and
Sarasua’s (2012) theories that women performed more paid labor than what is apparent in
much of the data we have, as well as their assertions that women participated in casual labor
in similar ways to men.
As mentioned before, one of the strengths of the data from Kalmar is that virtually every
individual is named, which can give some strong indications of women’s occupational roles
that are not available in other sources. Women are typically included under the heading
‘hantlangare’, as mason assistants and unskilled workers, along with men. While female
builders are often listed at the end of daily lists of workers in the Malmo data, men and
women are thoroughly mixed in Kalmar, with individuals who worked the most days listed
first; either Anna Dikerska or Anna Kalkslagerska top almost every list of unskilled workers
throughout the entire period.
Anna Kalkslagerska and Anna Dikerska are also notable because of their names; both
women’s surnames are occupational titles, rather than the more typical patronyms that most
women and many men carried. Many master men adopted the names of their professions, of
course; there are many men called ‘Murmästare’ (mason) and ‘Timberman’ (carpenter), along
with a few named ‘Dödgravare’ (gravedigger), but it is in general less typical for women to
have occupational names. The women’s titles are less prestigious than the men’s;
‘Kalkslagerska’ is a mortar mixer or plasterer, and ‘Dikerska’ is a digger, but the suffix ‘ska’
unambiguously indicate that these titles are female, and so were not inherited from a father or
husband; instead these names must refer to these women’s identification as professional
construction workers.
Swedish women in an international perspective
As mentioned above, there is a substantial lack of data on women’s historical wages, which
makes an international comparison difficult. Here, Swedish women’s construction wages are
compared with women’s causal wages in England (Humphries and Weisdorf 2015) and the
wages of women working at Winchester College in England (van Zanden 2011).
It is important to note that the wages from both English sources represent different types of
work than those that come from Sweden; while the Swedish women are essentially all
construction workers, the women in Humphries and Weisdorf’s series are performing a large
number of unskilled tasks, and women at Winchester College performed more domestic and
agricultural manual labor, such as cleaning and winnowing. It is also important to put these
series into context; Sweden was a largely rural and agricultural society until the end of the
16
K. E. Gary
eighteenth century, while England was an early industrializer, with mechanization changing
many industries already at the end of the eighteenth century. These important differences
make a comparison difficult, but it is still worthwhile to see where Sweden fits compared to
others.
The development of men’s and women’s relative payments in Sweden is contrary to the
development of English wage ratios. Humphries and Weisfdorf (2015) find a declining level
of relative casual wages from the middle of the sixteenth century; while there was some
recovery through the seventeenth century, the decline appeared again in the middle of the
eighteenth century, as the introduction of mechanized spinning drastically undercut spinners’
opportunities for income and changed the overall casual labor market for women.
At the beginning of the sixteenth century these women’s wages were on par with the unskilled
men also hired by the college, but a substantial wage gap quickly developed, as men’s wages
grew but women’s remained low; by the middle of the seventeenth century women were
earning less than half of what men could earn. Like in Humphries and Weisdorf’s aggregate
series, there was substantial wage recovery for women in the later seventeenth century,
bringing them almost to seventy percent of men’s wages, but still not enough to bring wages
back to parity.
Figure 5: women’s realtive wages in England and Sweden
Source: Winchester College: van Zanden 2011. England: Humphries and Weisdorf 2015. Sweden:
see text.
0
0,2
0,4
0,6
0,8
1
1,21
50
0
15
10
15
20
15
30
15
40
15
50
15
60
15
70
15
80
15
90
16
00
16
10
16
20
16
30
16
40
16
50
16
60
16
70
16
80
16
90
17
00
17
10
17
20
17
30
17
40
17
50
17
60
17
70
17
80
17
90
18
00
Relative wages Winchester College, UK (vZ 2011)
Relative wages in England (H&W 2015)
Relative wages in Sweden
17
K. E. Gary
This pattern of a steep fall and decline, followed by a tentative recovery, is also the dominant
trend in the Swedish wages, but the pattern begins much earlier in the English data, already at
the middle of the sixteenth century; the Scandinavian trend is about fifty years behind the
English development. Additionally, these developments are likely connected to different
trends. Humphries and Weisdorf connect the decline in women’s relative unskilled wages in
the eighteenth century to the increase in competition women faced from mechanization, while
the high relative wage levels in Sweden are more likely connected to periods of high labor
demand connected to destructive warfare; mechanization entered the scene relatively late in
Sweden, and so is less likely to have had a strong impact during this period.
Discussion and conclusion
This study has shown that not only did women work extensively in the early modern Swedish
construction industry, but that they were able to command high wages when labor demand
was high, and entered construction work in response to demand factors. The combination of
the potential for women to earn wages on par with their male peers in construction, a physical
industry, argues against women being less productive workers than men, or at least against a
payment structure based on such a relationship, as in Burnette (2008). At the same time, high
earning periods alongside periods of decline argue against pure discrimination, as in Bardsley
(1999), though it does indicate some reversion to a discriminatory state after the building
crises were over, in which women were not the preferred workers. When physical labor was
needed women were not excluded or underpaid, especially when recorded in the payrolls
directly with men – women’s relative wages are often lower when they were worked without
men, indicating women yielded a certain amount of bargaining power in periods of high work
demand, though it was unsustained. Together this indicates a limited agency due to
demographic pressures and labor needs, as theorized to have happened after the Black Death
(van Zanden 2011). Women’s response to demand factors also enrichens a literature which
tends to focus on the supply side of women’s labor in the pre-industrial Europe (see
Humphries and Weisdorf 2015; van Zanden 2011), and urges us to continue investigating
historical labor markets in deeper detail.
The wage gap in Swedish construction in the early modern period appears to be primarily
motivated by demand factors; as Humphries and Sarasua (2012) note, this aspect has not been
investigated as much as supply factors in investigating women’s participation in paid work in
the past. In the case of Sweden we see that when there is a high demand for construction work
employers make less distinction between male and female workers, both in terms of how they
are recorded and the rates at which they are payed. It is when levels of construction work are
lower and people are employed less regularly that wages begin to show some differentiation.
Women were responsive to demand factors, and this period of higher wages coincided with
both higher periods of labor demand and higher levels of female employment. This pattern
shifted after the beginning of the seventeenth century, when women’s nominal wages fell
18
K. E. Gary
substantially while men’s generally continued to increase; larger building projects became
less frequent, and when they did occur they employed fewer women.
Theory would predict that when we observe women performing strength-intensive jobs that
we would expect these women to earn more than other women, because the women working
in strength-intensive jobs must be able to earn more in these occupations than they would in
occupations that did not require strength in order to make the occupational change. However,
theory still predicts that women in physical occupations will earn less than men in the same
professions (Burnette 2008). This is not always the case in the data here, especially in the data
from Kalmar where men and women working as unskilled workers are listed together and
paid the same day rate; the heterogeneity in wages largely appears when there is lower
demand for labor and tasks are less regularly paid. However, a similar sorting-based
mechanism could explain why women tend to work more than unskilled men; if only those
women who are most able are regularly obtaining employment in construction work, it is
possible they are particularly efficient and choose to remain in this industry, while men are
find other employment. Alternatively, it could be a function of a lack of employment
opportunities for women compared to men.
The decline of women’s relative wages is concomitant with a decline in building demand, as
well as with a decline in the relative proportion of women working in construction. This
makes it unlikely that women’s relative employment is due to any sort of substitution between
unskilled men and women; instead, they seem to act as compliments. Similarly, the rapid
shifts in relative pay levels within the same sources, or at least the same geographical regions,
also undercut a story based on customary gendered pay gaps; even if the differences in
gendered pay rates are due to discrimination, the patterns of discrimination are not systematic
or entrenched across time.
Regardless of what is causing the changes in relative payments it is clear that women are
separated into the lowest class of workers; the fact that they are at times paid more or less the
same as the other unskilled men doesn’t change the fact that women were not able to become
more than minimally skilled; they were still severely limited in their working options.
Women were certainly not strangers to paid physical labor; evidence from Kalmar shows that
they were frequently between thirty and sixty percent of the unskilled labor force on
construction projects, and records from Malmo indicate that women are ‘invisible’ in the data
in many instances, and were likely working alongside men for equal or similar pay more
frequently that we are able to observe.
19
K. E. Gary
Primary data sources
Lunds Landsarkiv (Regional Archives in Lund)
Urban archives:
Landskrona rådhusrätt och magistrat (Landskrona city court and magistrate)
Lunds domkyrkas arkiv (Lund Cathedral archive)
Lunds stadsarkiv: Rådhusrättens och magistratens arkiv (Lund city archive: City Hall Court
and magistrate archive)
Ystad stadsarkiv: Rådhusrättens och magistratens arkiv (Ystad city archive: City Hall Court
and magistrate archive)
Manorial archives:
Jordberga godsarkiv (Jordberga estate archive)
Karsholms godsarkiv (Karsholm estate archive)
Knutstorps godsarkiv (Knutstorp estate archive)
Maltesholms godsarkiv (Maltesholm estate archive)
Rosendals godsarkiv (Rosendal estate archive)
Rydsgårds godsarkiv (Rydsgård estate archive)
Trolle Ljungby godsarkiv (Trolle Ljungby estate archive)
Vittskövle godsarkiv (Vittskövle estate archive)
Malmö Stadsarkiv (Malmo City Archives)
Urban archives:
Borgerskapet i Malmö 1517 – 1862 (Burghers in Malmo 1517-1862)
Malmö Hospital 1528 – 1923 (Malmo Hospital 1528-1923)
S:t Petri kyrkoarkiv (Saint Petri church archive)
Landsarkivet i Vadstena (Regional Archives in Vadstena)
Kalmar rådhusrätt och magistrat 1600-1850 (Kalmar city court and magistrate 1600-1850)
Secondary data sources
Albildgren, K., 2010. “Consumer Prices in Denmark 1502-2007.” Scandinavian Economic
History Review, 58(4), pp. 2-24.
20
K. E. Gary
Edvinsson, R., and Söderberg, J., 2007. A Consumer Price Index for Sweden 1290-2006.
[Excel spreadsheet]. Svergies Riksbank. Available at:
http://www.riksbank.se/en/The-Riksbank/Research/Historical-Monetary-
Statistics-/Prices/
Jansson, A.; Palm, L. A.; and Söderberg, J., 1991, Dagligt bröd i onda tider : priser och löner
i Stockholm och Västsverige 1500-1770, Göteborg: Institutet för lokalhistorisk
forskning.
Jörberg, L, 1972, A history of prices in Sweden 1732-1914. Vol. 1, Sources, methods, tables,
Lund: Gleerup.
Söderberg, J., 2002. Prices in Stockholm 1539-1620. [Excel spreadsheet]. Available at:
gpih.ucdavis.edu/files/Stockholm_1539-1620.xls
Söderberg, J., 2007. Daily wages of unskilled labourers in Stockholm 1540-1850. [Excel
spreadsheet].Sveriges Riksbank. Available at: http://www.riksbank.se/en/The-
Riksbank/Research/Historical-Monetary-Statistics-/Wages/Daily-wages-of-
unskilled-labourers-in-Stockholm-1501-1850/
Söderberg, J. Prices and wages in 16th century Stockholm. Unpublished notes. Received
October 2014.
Tomner, L., 1964. Forskningar I Malmö ekonomiska historia före 1658, unpublished
licentiate thesis, Lund: Department of Economic History.
References
Allen, R. C., 2001. “The Great Divergence in European Wages and Prices from the Middle
Ages to the First World War.” Explorations in Economic History 38, pp. 411-
447.
Allen, R. C., and Weisdorf, J. L., 2011. “Was there and ‘industrious revolution’ before the
industrial revolution? An empirical exercise for England c. 1300-1830. The
Economic History Review 64(3), pp. 715-729
Andersson-Palm, L., 2000. Folkmängden I Sveriges socknar och kommuner 1571-1997.
Stockholm: Nykopia Tryck AB.
Bardsley, S., 1999. “Women’s work reconsidered. Gender and wage differentiation in late
medieval England.” Past and Present 165, pp. 3-29.
Bardsley, S., 2001.[ Women’s work reconsidered. Gender and wage differentiation in late
medieval England]: Reply. Past and Present 173, pp. 199-202.
Burnette, J., 2004. “The wages and employment of female day-laboureres in English
agriculture, 1740-1850.” Economic History Review 57(4), pp. 664-690.
21
K. E. Gary
Burnette, J., 2008. Gender, Work, and Wages in Industrial Revolution Britain. Cambridge:
Cambridge University Press.
de Vries, J., 1994. “The Industrial Revolution and the Industrious Revolution.” The Journal of
Economic History 54(2), pp. 249-270.
de Vries, J., 2008. The Industrious Revolution. Cambridge: Cambridge University Press.
Edvinsson, R., 2010. “The multiple currencies of Sweden-Finalnd 1534-1803.” In: Edvinsson,
R., Jacobson, T., and Waldenström, D., eds. Historical Monetary and Financial
Statistics for Sweden: Exchange rates, prices, and wages 1277-2008. pp 453-
478. Sveriges Riksbank and Ekerlids Förlag: Halmstad.
Friis, A., and Glasmann, K., 1958. A History of Prices and Wages in Denmark: 1660-1800.
Vol 1. London: Longmans, Green and Co.
Gary, K., 2014. “Wages in comparison. Scandinavia in the early modern period: Evidence
from Scania.” Master’s thesis, Lund University.
Goldin, C., 1994. “The U-Shaped Female Labor Force Function in Economic Development
and Economic History.” NBER Working Paper no. 4707.
de Moor, T., and van Zanden, J. L., 2010. “Girl power: the European marriage pattern and
labour markets in the North Sea region in the late medieval and early modern
period.” Economic History Review 63(1)., pp 1-33.
Humphries, J., and Sarasua, C., 2012. “Off the Record: Reconstructing Women’s Labor Force
Participation in the European Past.” Feminist Economics, 18(4)
Humphries, J., and Weisdorf, J., 2015. “The Wages of Women in England, 1260-1850.”
Journal of Economic History 75, pp. 405-447.
Leeuwen, M. D. H. van, Maas, I., Miles, A., and Edvinsson, S., 2002. HISCO: Historical
International Standard Classification of Occupations. Leuven: Leuven
University Press.
Leeuwen, M. D. H. van, Maas, I, 2011. HISCLASS: A Historical International Social Class
Scheme. Leuven: Leuven University Press.
van Zanden, J. L., 2011. “The Malthusian Intermezzo: Women’s wages and human capital
formation between the late Middle Ages and the demographic transition of the
19th century.” History of the Family 16, pp. 331-342.
22
K. E. Gary
Appendix table 1: Archival sources
Archive name Wage payments (N)
Women (N) Workdays (N) (rounded to nearest workday)
Date range Type of archive
Malmö borgerskapet
13,063 336 33,895 1587-1759 Town
Kristianstad 69 0 1,157 1659-1727 Town
Lunds rådhusrätt
55 0 208 1674-1757 Town
Halmstad 44 0 224 1594-1702 Town
Ystad rådhuset
28 0 1636-1737 Town
Vadstena rådhusrätt och magistrat
25 3 335 1613-1729 Town
Landskrona rådhusrätt
18 0 52 1699-1758 Town
Kalmar 2,417 1,333 4,551 1614-1709 Urban church
Malmö Petrikyrkan
627 1 4,137 1550-1681 Urban church
Malmö hospital
514 14 1,787 1550-1676 Urban hospital
Allmäna barnhuset
239 2 4,385 1642-1643 Urban orphanage
Lunds domkyrka
182 0 928 1683-1759 Urban church (cathedral)
Domkapitlet i Växjö
56 7 281 1659-1703 Urban church (cathedral)
Vittskövle 274 3 4,603 1695-1759 Manor
Trolle Ljungby 203 1 1,241 1721-1758 Manor
Jordberga 31 0 128 1756-1759 Manor
Maltesholm 9 0 64 1738-1756 Manor
Karsholm 1 0 19 1707 Manor
Total 17,855 1,700 57,995 1550-1759